SAGA-GIS Module Library Documentation (v2.2.2)

Module Multiple Regression Analysis (Points and Predictor Grids)

Linear regression analysis of point attributes with multiple grids. Details of the regression/correlation analysis will be saved to a table. The regression function is used to create a new grid with regression based values. The multiple regression analysis uses a forward selection procedure.

Reference:
- Bahrenberg, G., Giese, E., Nipper, J. (1992): 'Statistische Methoden in der Geographie 2 - Multivariate Statistik', Stuttgart, 415p.

Parameters

 NameTypeIdentifierDescriptionConstraints
InputPredictorsGrid list (input)PREDICTORS--
PointsShapes (input)POINTS--
OutputDetails: Coefficients (*)Table (optional output)INFO_COEFF--
Details: Model (*)Table (optional output)INFO_MODEL--
Details: Steps (*)Table (optional output)INFO_STEPS--
Residuals (*)Shapes (optional output)RESIDUALS--
RegressionGrid (output)REGRESSIONregression model applied to predictor grids-
Regression with Residual Correction (*)Grid (optional output)REGRESCORRregression model applied to predictor grids with interpolated residuals added-
OptionsDependent VariableTable fieldATTRIBUTE--
Grid InterpolationChoiceINTERPOL-Available Choices:
[0] Nearest Neighbor
[1] Bilinear Interpolation
[2] Inverse Distance Interpolation
[3] Bicubic Spline Interpolation
[4] B-Spline Interpolation
Default: 4
Include X CoordinateBooleanCOORD_X-Default: 0
Include Y CoordinateBooleanCOORD_Y-Default: 0
InterceptBooleanINTERCEPT-Default: 1
MethodChoiceMETHOD-Available Choices:
[0] include all
[1] forward
[2] backward
[3] stepwise
Default: 3
Significance LevelFloating pointP_VALUESignificance level (aka p-value) as threshold for automated predictor selection, given as percentageMinimum: 0.000000
Maximum: 100.000000
Default: 5.000000
Cross ValidationChoiceCROSSVAL-Available Choices:
[0] none
[1] leave one out
[2] 2-fold
[3] k-fold
Default: 0
Cross Validation SubsamplesIntegerCROSSVAL_Knumber of subsamples for k-fold cross validationMinimum: 2
Default: 10
Residual InterpolationChoiceRESIDUAL_COR-Available Choices:
[0] Multleve B-Spline Interpolation
[1] Inverse Distance Weighted
Default: 0
(*) optional

Command-line

Usage: saga_cmd statistics_regression 1 [-PREDICTORS <str>] [-POINTS <str>] [-ATTRIBUTE <str>] [-INFO_COEFF <str>] [-INFO_MODEL <str>] [-INFO_STEPS <str>] [-RESIDUALS <str>] [-REGRESSION <str>] [-REGRESCORR <str>] [-INTERPOL <str>] [-COORD_X <str>] [-COORD_Y <str>] [-INTERCEPT <str>] [-METHOD <str>] [-P_VALUE <str>] [-CROSSVAL <str>] [-CROSSVAL_K <num>] [-RESIDUAL_COR <str>]
  -PREDICTORS:<str>  	Predictors
	Grid list (input)
  -POINTS:<str>      	Points
	Shapes (input)
  -ATTRIBUTE:<str>   	Dependent Variable
	Table field
  -INFO_COEFF:<str>  	Details: Coefficients
	Table (optional output)
  -INFO_MODEL:<str>  	Details: Model
	Table (optional output)
  -INFO_STEPS:<str>  	Details: Steps
	Table (optional output)
  -RESIDUALS:<str>   	Residuals
	Shapes (optional output)
  -REGRESSION:<str>  	Regression
	Grid (output)
  -REGRESCORR:<str>  	Regression with Residual Correction
	Grid (optional output)
  -INTERPOL:<str>    	Grid Interpolation
	Choice
	Available Choices:
	[0] Nearest Neighbor
	[1] Bilinear Interpolation
	[2] Inverse Distance Interpolation
	[3] Bicubic Spline Interpolation
	[4] B-Spline Interpolation
	Default: 4
  -COORD_X:<str>     	Include X Coordinate
	Boolean
	Default: 0
  -COORD_Y:<str>     	Include Y Coordinate
	Boolean
	Default: 0
  -INTERCEPT:<str>   	Intercept
	Boolean
	Default: 1
  -METHOD:<str>      	Method
	Choice
	Available Choices:
	[0] include all
	[1] forward
	[2] backward
	[3] stepwise
	Default: 3
  -P_VALUE:<str>     	Significance Level
	Floating point
	Minimum: 0.000000
	Maximum: 100.000000
	Default: 5.000000
  -CROSSVAL:<str>    	Cross Validation
	Choice
	Available Choices:
	[0] none
	[1] leave one out
	[2] 2-fold
	[3] k-fold
	Default: 0
  -CROSSVAL_K:<num>  	Cross Validation Subsamples
	Integer
	Minimum: 2
	Default: 10
  -RESIDUAL_COR:<str>	Residual Interpolation
	Choice
	Available Choices:
	[0] Multleve B-Spline Interpolation
	[1] Inverse Distance Weighted
	Default: 0